Hereditary hemochromatosis (HH) is an inherited disorder of iron metabolism wherein the body accumulates excess iron. In symptomatic individuals, this excess iron leads to deleterious effects by being deposited in a variety of organs leading to their failure, and resulting in cirrhosis, diabetes, sterility, and other serious illnesses. Neither the precise physiological mechanism of iron overaccumulation nor the gene which is defective in this disease has been described.
HH is typically inherited as a recessive trait; in the current state of knowledge, homozygotes carrying two defective copies of the gene are most frequently affected by the disease. In addition, heterozygotes for the HH gene are more susceptible to sporadic porphyria cutanea tarda and potential other disorders (Roberts et al., Lancet 349:321-323 (1997). It is estimated that approximately 10-15% of individuals of Western European descent carry one copy of the HH gene mutation and that there are about one million homozygotes in the United States. HH, thus, represents one of the most common genetic disease mutations in individuals of Western European descent. Although ultimately HH produces debilitating symptoms, the majority of homozygotes and heterozygotes have not been diagnosed.
The symptoms of HH are often similar to those of other conditions, and the severe effects of the disease often do not appear immediately. Accordingly, it would be desirable to provide a method to identify persons who may be destined to become symptomatic in order to intervene in time to prevent excessive tissue damage associated with iron overload. One reason for the lack of early diagnosis is the inadequacy of presently available diagnostic methods to ascertain which individuals are at risk, especially while such individuals are presymptomatic.
Although blood iron parameters can be used as a screening tool, a confirmed diagnosis often employs liver biopsy which is undesirably invasive, costly, and carries a risk of mortality. Thus, there is a clear need for the development of an inexpensive and noninvasive diagnostic test for detection of homozygotes and heterozygotes in order to facilitate diagnosis in symptomatic individuals, provide presymptomatic detection to guide intervention in order to prevent organ damage, and for identification of heterozygote carriers.
The need for such diagnostics is documented, for example, in Barton, J. C. et al. Nature Medicine 2:394-395 (1996); Finch, C. A. West J Med 153:323-325 (1990); McCusick, V. Mendelian Inheritance in Man pp. 1882-1887, 11th ed., (Johns Hopkins University Press, Baltimore (1994)); Report of a Joint World Health Organization/Hemochromatosis Foundation/French Hemochromatosis Association Meeting on the Prevention and Control of Hemochromatosis (1993); Edwards, C. Q. et al. New Engl J Med 328:1616-1620 (1993); Bacon, B. R. New Engl J Med 326:126-127 (1992); Balan, V. et al. Gastroenterology 107:453-459 (1994); Phatak, P. D. et al. Arch Int Med 154:769-776 (1994).
Although the gene carrying the mutation or mutations that cause HH has previously been unknown, genetic linkage studies in HH families have shown that the gene that causes the disease in Caucasians appears to reside on chromosome 6 near the HLA region at 6p21.3 (Cartwright, Trans Assoc Am Phys 91:273-281 (1978); Lipinski, M. et al. Tissue Antigens 11:471-474 (1978)). It is believed that within this locus, a single mutation gave rise to the majority of disease-causing chromosomes present in the population today. See Simon, M. et al. Gut 17:332-334 (1976); McCusick, V. supra. This is referred to herein as the "common" or "ancestral" or "common ancestral" mutation. These terms are used interchangeably. It appears that about 80% to 90% of all HH patients carry at least one copy of the common ancestral mutation which is closely linked to specific alleles of certain genetic markers close to this ancestral HH gene defect. These markers are, as a first approximation, in the allelic form in which they were present at the time the ancestral HH mutation occurred. See, for example, Simon, M. et al. Am J Hum Genet 41:89-105 (1987); Jazwinska, E. C. et al. Am J Hum Genet 53:242-257 (1993); Jazwinska, E. C. et al. Am J Hum Genet 56:428-433 (1995); Worwood, M. et al. Brit J Hematol 86:863-866 (1994); Summers, K.M. et al. Am J Hum Genet 45:41-48 (1989).
Several polymorphic markers in the putative HH region have been described and shown to have alleles that are associated with HH disease. These markers include the published microsatellite markers D6S258, D6S306 (Gyapay, G. et al. Nature Genetics 7:246-339 (1994)), D6S265 (Worwood, M. et al. Brit J Hematol 86:833-846 (1994)), D6S105 (Jazwinska, E. C. et al. Am J Hum Genet 53:242-257 (1993); Jazwinska, E. C. et al. Am J Hum Genet 56:428-433 (1995)), D6S1001 (Stone, C. et al. Hum Molec Genet 3:2043-2046 (1994)), D6S1260 (Raha-Chowdhury et al. Hum Molec Genet 4:1869-1874 (1995)) as well as additional microsatellite and single-nucleotide-polymorphism markers disclosed in co-pending PCT application WO 96/35802 published Nov. 14, 1996, the disclosure of which is hereby incorporated by reference in its entirety.
Although each of such markers may be of use in identifying individuals carrying the defective HH gene, crossing-over events have, over time, separated some of the ancestral alleles from the mutation that is responsible for HH, thereby limiting the utility of such surrogate markers. The limited diagnostic power of surrogate markers is obvious considering the fact that the frequency of the ancestral allele in the population is generally higher than the estimated frequency of the disease-causing mutation. The desirability of identifying the gene responsible for HH has long been recognized due to the health benefits that would be available via gene-based diagnostics, which has an intrinsically higher predictive power than surrogate markers and may eventually lead to the identification and diagnosis of disease-causing mutations other than the ancestral mutation. In addition, identification of the HH gene would further our understanding of the molecular mechanisms involved in HH disease thereby opening new approaches for therapy. This goal has motivated numerous, but previously unsuccessful attempts to identify the HH gene.
These attempts have been made by a variety of methods. For example, genes known to be involved in iron transport or metabolism have been examined as candidates. An example of one unsuccessful attempt is the assignment of the ferritin heavy chain gene to Chromosome 6p, and subsequent exclusion of this gene on the basis of its precise localization outside of the HH region, and failure to find mutations in HH patients. See Dugast, I. J. et al. Genomics 6:204-211 (1990); Summers et al. Hum Genet 88:175-178 (1991).
Another strategy has been to employ the genomic DNA surrounding the postulated HH locus to select expressed genes from this region. These genes have been evaluated in HH patients for mutations in an attempt to identify them as the causative gene. Examples of searches that have not resulted in the identification of the HH gene are illustrated in El Kahloun et al. Hum Molec Genet 2:55-60 (1992), Goei et al. Am J Hum Genet 54:244-251 (1994), and Beutler et al. Blood Cells, Molecules, and Diseases 21:206-216 (1995).
Finally, although the strategy of using positional information obtained from genetic studies has long been a widely used approach, estimates of the position of the HH gene remained imprecise. Examples of this uncertainty are demonstrated in Gruen et al. Genomics 14:232-240 (1992) and in Gasparini et al. Hematology 19:1050-1056 (1994). Indeed, a number of contradictory conclusions have been reported, some placing the HH gene proximal of HLA-A (Edwards et al. Cytogenet Cell Genet 40:620 (1985); Gasparini, P. et al. Hum Molec Genet 2:571-576 (1993)) while others placed the gene distal of HLA-A (Calandro et al. Hum Genet 96:339-342 (1995)).
Until very recently, in spite of the linkage studies placing the HH disease gene in the HLA region of Chromosome 6, the biological relevance of alterations in HLA Class I components has not been particularly well explored. Work by de Sousa et al. Immun Lett 39:105-111 (1994), and more recent work by Rothenberg, B. E. and Voland, J. R. Proc Natl Acad Sci USA 93:1529-1534 (1996) indicated that .beta.-2-microglobulin knock-out mice develop symptoms of iron overload. .beta.-2-microglobulin is presented on cell surfaces as a complex with HLA Class I MHC's. de Sousa et al. supra. (1994) and Barton, J. C. and Bertoli, L. F. Nature Medicine 2:394-395 (1996) speculated that .beta.-2-microglobulin associated proteins or a unique Class I gene could be involved in the control of intestinal iron absorption and possibly HH disease.
In spite of the extensive efforts in the art to find the gene responsible for HH, the gene has remained elusive. Nevertheless, as will be appreciated it would be highly desirable to identify, isolate, clone, and sequence the gene responsible for HH and to have improved diagnostic methods for detection of affected individuals, whether homozygotes or heterozygotes.